Recently, Israel's Weizmann Institute of Science announced that its research team has developed an artificial intelligence model Brain-IT, which can reconstruct image content based on functional magnetic resonance imaging data when humans view images. Unlike existing methods, which usually require tens of hours of brain scan data, this model only takes about an hour to learn to decode the brain activity of new subjects, and can more accurately restore details such as image content, composition, and color. The research team mitigated the problem of insufficient training samples by developing image encoders and decoders, and using artificially generated predictive brain activity data. During the research, the encoder also identified 128 image processing functional areas common to different groups of people, some of which had not been discovered before. The research team said that Brain-IT related research has been selected for the International Learning Characterization Conference. Currently, the team is exploring extending this technology to decoding sound information and investigating future possibilities for video decoding.

Disclaimer:Webull uses external vendor Google Translation Service for news translations where we endeavour to ensure these are correct, however, we recommend that you please double-check this information accordingly. Webull is not responsible for translation errors or issues.